Methods and computing systems are disclosed for a digital assistant platform that includes a cross-platform bot that can be used to bridge existing chat-bot platforms. The cross-platform bot would have its own single set of protocols and standards, such that a developer of a task-specific bot would only need to integrate with the cross-platform bot to enable its use across all the different chat-bot platforms. The cross-platform bot would enable the task-specific bots to coordinate to combine their functionality, while end users may continue using an existing chat-bot platform of their choice.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A computer-implemented method, comprising: obtaining, from a chat-bot platform, a conversational command provided by a user, wherein the conversational command instructs performance of a task; determining whether the conversational command designates a task specific bot for performance of the task; upon determining that the conversational command cannot determine a single task specific bot for performance of the task, then analyzing an intent of the conversational command; determining that the intent of the conversational command indicates a plurality of actions; determining two or more task specific bots for performance of the plurality of actions associated with the task based on the intent of the conversational command and a bot registry, wherein the bot registry lists the two or more task specific bots and functionality associated with each of the two or more task specific bots and wherein at least one of the two or more specific task bots is selected from a plurality of task bots that can perform tasks based upon one of the plurality of actions associated with the task using a real-time bidding process; and providing the conversational command or portions thereof to the determined two or more task specific bots for performance of the task.
2. The method of claim 1 , wherein the two or more task specific bots are added to the bot registry by a developer of each respective task specific bot.
3. The method of claim 1 , wherein analyzing the intent of the conversational command includes referencing the intent of the conversational command in an intent-action map store.
4. The method of claim 1 , wherein the bot registry further lists the two or more task specific bots based on interactions conducted in a bot exchange.
5. The method of claim 1 , wherein the bot registry further includes interaction results associated with each of the two or more task specific bots, wherein the interaction results are conducted in a bot exchange.
6. The method of claim 1 , wherein the determined two or more task specific bots perform bot chaining for performance of the task.
7. The method of claim 6 , wherein bot chaining is managed by a bot policy server.
8. The method of claim 1 , wherein at least one of the determined two or more task specific bots is a value processing bot.
9. The method of claim 1 , wherein the conversational command is provided by the user as a voice command.
10. A computing system, comprising: one or more processors; non-transitory computer-readable memory storing instructions that, when executed by the one or more processors, cause the one or more processors to: obtain, from a chat-bot platform, a conversational command provided by a user, wherein the conversational command instructs performance of a task; determine whether the conversational command designates a task specific bot for performance of the task; upon determining that the conversational command cannot determine a single task specific bot for performance of the task, then analyzing an intent of the conversational command; determine that the intent of the conversational command indicates a plurality of actions; determine two or more task specific bots for performance of the plurality of actions associated with the task based on the intent of the conversational command and a bot registry, wherein the bot registry lists the two or more task specific bots and functionality associated with each of the one or more task specific bots and wherein at least one of the two or more specific task bots is selected from a plurality of task bots that can perform tasks based upon one of the plurality of actions associated with the task using a real-time bidding process; and provide the conversational command or portions thereof to the determined two or more task specific bots for performance of the task.
11. The computing system of claim 10 , wherein the two or more task specific bots are added to the bot registry by a developer of each respective task specific bot in the two or more task specific bots.
12. The computing system of claim 10 , wherein analyzing the intent of the conversational command includes referencing the intent of the conversational command in an intent-action map store.
13. The computing system of claim 10 , wherein the bot registry further lists the two or more task specific bots based on interactions conducted in a bot exchange.
14. The computing system of claim 10 , wherein the bot registry further includes interaction results associated with each of the two or more task specific bots, wherein the interaction results are conducted in a bot exchange.
15. The computing system of claim 10 , wherein the determined two or more task specific bots perform bot chaining for performance of the task.
16. The computing system of claim 15 , wherein a bot policy server manages bot chaining.
17. The computing system of claim 10 , wherein at least one of the determined two or more task specific bots is a value processing bot.
18. The computing system of claim 10 , wherein the conversational command is provided by the user as a voice command.
19. The method of claim 1 , wherein at least one of the two or more specific task bots is selected from a set of task bots in the plurality of task bots, wherein the tasks bots in the set of tasks bots can perform tasks based upon a same action.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
November 6, 2018
December 8, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.